How to use the AI readiness assessment
Pick one task before you start: chasing an invoice, writing up site notes or preparing a weekly report. Answer for that task, rather than for the whole business.
For each question record clear, uncertain or missing, and note the evidence behind your answer. Name a person to resolve each gap. This is a discussion tool, not a certification, risk assessment or guarantee that a project should proceed.
The AI readiness checklist
Use these 12 questions with the people who perform the task. Record the current process, rather than the way it is meant to work.
The work
- Can you name one task you want AI to help with?
- How often does that task happen?
- Who owns it, and are the steps written down?
- Do you know how long it takes today?
An AI project without a named task has nothing to measure. A task that happens twice a year rarely pays back the setup.
Information and systems
- Where does the information for this task live?
- Do you know whether the task uses personal or client-confidential information, and what rules apply?
- How are user accounts and file access managed?
- Are staff already using AI tools for work, and on which accounts?
AI tools work from the information they can reach. Microsoft says Microsoft Copilot (formerly Microsoft 365 Copilot) “can also use content that a user is permitted to access in archive, shared, and delegate mailboxes”. It strongly recommends a secure and governed foundation, including fixing oversharing. So if an old sharing link opens the HR folder to everyone, an assistant may bring it up. (Microsoft Copilot requirements, checked 25 October 2026.)
Personal data raises the bar. The ICO’s list of processing likely to result in high risk includes “the use of innovative technologies, or the novel application of existing technologies (including AI)”. It says a data protection impact assessment is required where this is combined with any of the criteria from the European guidelines. (ICO guidance on when a DPIA is needed, checked 25 October 2026.)
People and decisions
- Who would decide to go ahead?
- Is there a budget for a first project?
- How do the people who do the task feel about changing it?
- What happens today when something in this task goes wrong?
Someone has to say yes, and the people doing the work have to want the change. Question 12 asks whether anyone checks the work before it goes out. The agents we build have review points. If nobody checks the work today, build that habit first.
How to decide the next step
| What you found | A sensible next step |
|---|---|
| No named task, owner or baseline | Write down the process and measure a normal week first. |
| Uncertain data rules or access | Resolve permissions and data handling with the relevant owner before connecting a tool. |
| A recurring task, usable inputs and a reviewer | Consider a small pilot with a defined outcome and a way to stop it. |
| Several competing opportunities | A wider audit can help rank value, effort and risk. |
A serious gap matters more than a neat total score. Start with the AI policy template if staff do not know what data they may use. For Microsoft 365, use the security review guide with your administrator.
What our AI audit for business covers
The checklist helps you identify questions. The audit investigates the business case in more detail. For £750 we look at:
- How the work moves. We review your workflows, operations and team structure, and we talk to the people involved.
- Which tools fit. Honest recommendations on the AI tools that suit your business. Sometimes the better fix is a simpler automation or a clearer process. If so, we will say so.
- Where the value is. A ranked list of AI opportunities, ordered by impact and ease of implementation.
- What it is worth. Estimated time and cost savings for each recommendation, so you can build a business case.
- What could go wrong. The likely risks, including risks to your data.
What you get at the end
- A written report in a clear, shareable format your team can act on
- The best workflow to tackle first, with its expected value
- Quick wins you can start on
- Our recommendation on the next step: training, automation or a larger implementation project
The plan is yours. You can use it with us, give it to your own team or take it to another supplier. Buying the audit does not commit you to anything else.
How the £750 AI audit runs
- Free 15-minute call. Tell us the task, the systems involved and who owns the work. If an audit is not the right step, we will tell you.
- Timetable in writing. We agree it before work starts. Your written quote confirms what is included and how payment works.
- We dig in. Process review, conversations with your team and analysis. We work mostly remotely. In-person sessions can be agreed before work starts.
- You get the plan. A report with prioritised recommendations and estimated savings.
- Optional: we build it. If you want us to implement any of the recommendations, we quote that separately.
The timetable is agreed in writing before work starts.
We agree the information and access required during scoping. Do not send confidential files or credentials before the handling arrangements are agreed.
Where we need access to a system, we work with whoever owns it, including your IT provider.
Who our AI readiness assessment for SMEs is for
The checklist is free for anyone. The audit is built for:
- Business owners who know AI matters but are not sure where to start
- Operations managers looking to cut costs and save time
- Leadership teams who need a business case before they invest
We work across sectors. Our case studies include AI training for a veterinary software company and for an independent boarding school in Cumbria.
The audit is probably not your best first step if:
- You already know the workflow and want it built. Go to AI implementation services.
- Your main need is staff skills. See our AI training courses or Microsoft Copilot training.
- Your worry is who can see what in Microsoft 365. Start with a Microsoft 365 security audit.
- You are a small team and want one adviser from first idea to build and training. See AI consulting for small businesses.
What a useful recommendation looks like
A recommendation should state the problem, evidence, proposed change, owner, dependencies and measure of success. Estimated savings should show their assumptions.
For example, a hypothetical invoice-chasing pilot could measure time spent preparing reminders and the proportion needing correction. Those are test measures, not promised client results. Start with a small approved sample and decide what evidence would justify expanding it.
Price, VAT and funding
- £750, fixed. Our prices are in pounds sterling. Your written quote confirms whether VAT applies and how payment works, including for businesses in the Republic of Ireland.
- Software is extra. Subscriptions, platform fees, hosting and licences are not included.
- No grant promises. Invest NI lists “working with a private sector provider” as not eligible for its Innovation Vouchers (Invest NI, checked 25 October 2026). So a voucher cannot pay for our work. Our Northern Ireland funding guide lists the official pages to check.
See all our prices.
Not sure which gap to tackle first?
Book a free 15-minute call and bring the task you had in mind. We will give you a straight answer on whether the audit is worth £750 to you, or whether you should start somewhere else.
FAQ
Is an AI readiness assessment the same as an AI audit? They overlap. An AI readiness assessment checks whether your workflows, data, tools and team are ready for AI. The free checklist helps you explore those questions for one task. The £750 AI audit goes further: it ranks the opportunities, sets out the risks and estimated savings, and gives you a written plan.
How much does the AI audit cost? £750, fixed price, agreed before work starts. Prices are in pounds sterling and your written quote confirms whether VAT applies. Third-party software and licence costs are separate. The audit does not commit you to a larger project.
How long does the AI audit take? We agree the timetable in writing before work starts.
Do we need any technical knowledge? No. You need to know your business and the task. We handle the technical side and explain the choices in plain English.
What if the checklist reveals gaps? That is a useful result. It tells you what to fix first, usually choosing one task, writing down how it works or setting rules for what information staff can put into AI tools. You don’t need to buy anything to do that. When you have closed the gaps, work through the checklist again.
Can we use the audit plan with another supplier? Yes. The plan is yours. You can use it with Blue Canvas, give it to your own team or take it to another supplier.
.svg){kind=small}
.svg){kind=small}